Triple
T16026458
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Mazda3 |
E388728
|
entity |
| Predicate | notableTrim |
P25549
|
FINISHED |
| Object | Mazda3 Turbo |
E388728
|
NE FINISHED |
How this triple was built (2 steps)
Every LLM step that produced this triple, in pipeline order — named-entity classification, the disambiguation choices (the exact options shown, with the pick highlighted), and the generated description. The batch + timestamp of each is in the Provenance table below.
NER
Named-entity recognition
gpt-5-mini
Instruction
Given a phrase, classify it is english named entity (e.g., persons, organizations, works of art) in Latin script, or not (e.g., literals, dates, URLs, verbose phrases). For disambiguation, the statement where the phrase occurs as object is also given. Please return a JSON object with `phrase` (string, the phrase being analyzed) and `is_ne` (boolean, indicating whether the phrase is a Named Entity).
Input
Phrase: Mazda3 Turbo | Statement: [Mazda3, notableTrim, Mazda3 Turbo]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Mazda3 Turbo Context triple: [Mazda3, notableTrim, Mazda3 Turbo]
-
A.
Mazda3
chosen
The Mazda3 is a popular compact car known for its sporty handling, stylish design, and well-appointed interior.
-
B.
Mazda6
The Mazda6 is a mid-size family sedan known for its sporty handling, stylish design, and strong value in the mainstream car market.
-
C.
Mazda RX-8
The Mazda RX-8 is a four-door sports coupe renowned for its lightweight chassis, distinctive freestyle rear doors, and high-revving rotary engine.
-
D.
Mazda2
The Mazda2 is a subcompact car known for its agile handling, fuel efficiency, and stylish design, positioned as an affordable entry-level model in Mazda’s lineup.
-
E.
Mazda SC
Mazda SC was a Japanese company football club owned by Mazda that later evolved into the professional J.League team Sanfrecce Hiroshima.
- F. None of above.
- G. Unsure - the case is ambiguous/there is not enough information to decide.
Provenance (3 batches)
The batch behind each pipeline step, in order, with when it ran. Timestamps are batch-level — stages were processed in waves, so the object chain (NER → NED1 → NEDg → NED2) reads in order, but predicate / elicitation batches can sit in a different wave.
| Step | Stage | Batch ID | Status | When |
|---|---|---|---|---|
| creating | Elicitation | batch_69d86dada3808190825d5f80d72fbe88 |
completed | April 10, 2026, 3:25 a.m. |
| NER | Named-entity recognition | batch_69e18328707c8190b9a444c78faaaa04 |
completed | April 17, 2026, 12:47 a.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69ffdbcfd39c81909bfddfe95f9ad7d2 |
completed | May 10, 2026, 1:13 a.m. |
Created at: April 10, 2026, 4:56 a.m.